Induction of Fatty Acid Synthesis Is a Key Requirement for Phagocytic Differentiation of Human Monocytes

Induction of fatty acid synthesis is a key requirement for phagocytic differentiation of human monocytes

Josef Eckera,1, Gerhard Liebischa,1, Marion Englmaiera,1, Margot Grandla, Horst Robenekb, and Gerd Schmitza,2

aInstitute of Clinical Chemistry and Laboratory Medicine, University of Regensburg, 93042 Regensburg, Germany; and bLeibniz Institute for Arteriosclerosis Research, University of Münster, 48149 Münster, Germany

Edited by David W. Russell, University of Texas Southwestern Medical Center, Dallas, TX, and approved March 19, 2010 (received for review October 20, 2009) Monocytes are precursors of macrophages. Here we demonstrate that onstrate an essential role of FA and phospholipid synthesis for the macrophage colony-stimulating factor (M-CSF)-dependent differentia- morphogenic alterations during phagocytic differentiation of mon- tion of primary human monocytes from healthy volunteers induces ocytes, including organelle development and macrophage function. transcription of SREBP-1c target genes required for fatty acid (FA) biosynthesis and impairs transcription of SREBP-2 target genes required Results for cholesterol synthesis. Detailed lipid metabolic profiling showed that Transcriptional Activation of SREBP-1c Target Genes and Inverse this transcriptional regulation leads to a dramatically increased fatty Regulation of SREBP-2 Target Genes During Monocyte Differentiation. acid synthesis as driving force for enhanced phospholipid synthesis. Global gene expression analysis of primary human M-CSF differ- During cell differentiation the major lipid class switches from cholesterol entiated monocytes from healthy donors with DNA microarrays in monocytes to phosphatidylcholine in macrophages. Ultrastructural revealed a significant induction of SREBP-1c and a down-regu- analysis revealed that this transcriptional and metabolic regulation is lation of SREBP-2 target gene expression (Table S1). To validate essential for development of macrophage filopodia and cellular or- these results, gene expression was analyzed with qRT-PCR of ganelles including primary lysosomes, endoplasmic reticulum, and Golgi undifferentiated monocytes and for 1, 4, or 6 days with M-CSF network. Additional functional studies showed that suppression of differentiated monocytes. We could verify the induction of SREBP- fatty acid synthesis prevents phagocytosis representing a central 1c regulated genes during monocyte differentiation (Table 1). FAS, macrophage function. Therefore induction of fatty acid synthesis is a ELOVL6, and SCD were massively up-regulated when M-CSF- key requirement for phagocyte development and function. dependent cell differentiation was induced. In sharp contrast, expression of genes involved in cholesterol metabolism remained fatty acid metabolism | lipid mass spectrometry | phagocytes | phospholipid unchanged or was only moderately increased when monocyte dif- synthesis ferentiation was induced. During later differentiation stages transcription of SREBP-2 target genes was down-regulated (except for acrophages are multifunctional cells of the innate immune PMVK, LSS, and DHCR7). Msystem present in all tissues in the body. They participate in numerous biological processes ranging from tissue and organ Induced Fatty Acid Synthesis and Desaturation During Monocyte development to bone remodelling and wound healing (1). A Differentiation. To test whether up-regulation of SREBP-1c target genes has an impact on macrophage lipid composition, FA profiles hallmark of macrophage function is their phagocytic capacity (2). A Differentiation of monocytes into macrophages is initiated by were analyzed during monocyte differentiation (Fig. 1 ). At day 4 macrophage colony-stimulating factor (M-CSF). M-CSF (also of M-CSF-mediated differentiation, we could detect a striking shift known as CSF-1) circulates at nanomolar levels in plasma and is of FA composition from saturated and polyunsaturated to mon- constitutively generated by several cell types. M-CSF levels are ounsaturated FAs. The C16 and C18 monounsaturated FA con- fl tent increased from 15% in monocytes to 38% in macrophages. elevated at sites of in ammation in pathological states including fi autoimmune disorders, vasculitis, arthritis, and obesity (3). In a next step, we asked whether the changes in FA pro les are due to increased activities of the enzymes FAS, SCD, and ELOVL6 Investigation ofmechanisms underlying monocyte differentiation fi is central to the understanding of fundamental macrophage biology (Fig. S1). To pro le FA synthesis, cells were incubated with stable and metabolic diseases. Although a link between peroxisome pro- isotope-labeled acetate and incorporation into palmitate was monitored (Fig. S1). In monocytes we observed almost no FA liferator-activated receptor-γ (PPARγ)-dependent inflammation synthesis, whereas induction of M-CSF-dependent differentiation and monocyte differentiation has been found (4, 5), little is known stimulated palmitate synthesis 28-fold (Fig. 1B). To explore FA about the role of lipid synthesis and almost none about the role of desaturation and elongation, cells were supplied with stable iso- fatty acid (FA) metabolism for monocyte differentiation. tope-labeled palmitate and conversion to palmitoleate and stearate CELL BIOLOGY Major regulators of lipid homeostasis in mammalian cells are the was determined (Fig. S1). Palmitate desaturation was not detect- nuclear transcription factors sterol regulatory element-binding able in monocytes, but in macrophages (Fig. 1C). Desaturation proteins (SREBPs), which belong to the family of basic helix loop ratios increased 25-fold from day 1 to day 6 of differentiation, helix leucine zipper (bHLH-LZ) transcription factors. SREBP-2 whereas palmitate elongation revealed only a modest rise during mainly controls genes of the cholesterol pathway and SREBP-1c monocyte differentiation (Fig. 1D). Taken together these lipid preferentially regulates genes involved in FA biosynthesis (6). A well-characterized SREBP-1c target gene and key enzyme required for endogenous FA synthesis in mammalian cells is fatty acid syn- Author contributions: J.E., G.L., and G.S. designed research; J.E., G.L., M.E., M.G., and H.R. thase (FAS), because it catalyzes the generation of palmitate from performed research; G.S. contributed new reagents/analytic tools; J.E., G.L., M.E., M.G., acetyl-CoA (7). Palmitate is either elongated to stearate by long and H.R. analyzed data; and J.E. and G.L. wrote the paper. chain fatty acid elongase 6 (ELOVL6) (8) or desaturated to pal- The authors declare no conflict of interest. mitoleate by stearoyl-CoA desaturase (SCD) (9). Fatty acid desa- This article is a PNAS Direct Submission. turase 1 (FADS1) and FADS 2 are SREBP-1c-regulated 1J.E., G.L., and M.E. contributed equally to this work. desaturases for polyunsaturated fatty acids (PUFAs) (10). 2To whom correspondence should be addressed. E-mail: [email protected] In this study we identified a massive up-regulation of genes regensburg.de. – required for FA synthesis during M-CSF dependent differentiation This article contains supporting information online at www.pnas.org/cgi/content/full/ of primary monocytes from healthy volunteers. We further dem- 0912059107/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.0912059107 PNAS | April 27, 2010 | vol. 107 | no. 17 | 7817–7822 Downloaded by guest on September 27, 2021 Table 1. Up-regulation of SREBP-1c target genes and down-regulation of SREBP-2 target genes during monocyte differentiation Gene/pathway Symbol Mean change Mean change Mean change

SREBP-1c target genes: fatty acid synthesis Mac(d1) Mac(d4) Mac(d6) Sterol regulatory element binding protein 1 SREBP-1c 1.36 8.38 6.43 Acetyl-CoA carboxylase alpha ACACA 0.97 1.26 2.01 Fatty acid synthase FAS 3.05 11.32 4.42 Elongase of long chain fatty acids 6 ELOVL6 12.27 131.04 74.72 Stearoyl-CoA desaturase 1 SCD1 3.95 58.42 39.26 Fatty acid desaturase 1 FADS1 1.42 6.29 2.75 Fatty acid desaturase 2 FADS2 2.48 12.93 6.72 SREBP-2 target genes: cholesterol synthesis Sterol regulatory element binding protein 2 SREBP-2 1.33 1.25 0.67 Acetyl-CoA acetyltransferase 2 ACAT2 1.56 1.87 0.87 3-Hydroxy-3-methylglutaryl-CoA reductase HMGCR 1.96 1.10 0.41 3-Hydroxy-3-methylglutaryl-CoA synthase 1 HMGCS1 1.77 0.73 0.31 Mevalonate kinase MVK 1.13 1.12 0.35 Phosphomevalonate kinase PMVK 1.26 1.68 1.44 Mevalonate decarboxylase MVD 0.96 0.69 0.38 Geranylgeranyl diphosphate synthase 1 GGPS1 0.7 0.62 0.62 Isopentenyl-diphosphate delta isomerase 1 IDI1 1.72 1.08 0.52 Farnesyl-diphosphate farnesyltransferase 1 FDFT1 1.22 0.85 0.29 Squalene epoxidase SQLE 1.95 1.29 0.55 Lanosterol synthase LSS 1.99 5.03 3.26 Cytochrome P450 (51A1) CYP51A1 1.67 1.20 0.57 Sterol-C5-desaturase SC5DL 1.73 1.33 0.83 7-dehydrocholesterol reductase DHCR7 4.19 11.52 2.72 Low density lipoprotein receptor LDLR 1.57 0.83 0.35

Expression of SREBP-1c and SREBP-2 target genes in macrophages differentiated with M-CSF for 1, 4, or 6 days compared with monocytes, analyzed with qRT-PCR; 18S rRNA was used as reference gene. The boldface type indicates that the change is ≥2or≤0.5.

species data are in good agreement with the transcriptional results. species during phagocytic differentiation of monocytes is in good The shift to monounsaturated C16 and C18 FAs in macrophages agreement with up-regulation of FA synthesis and desaturation results from a strong induction of FA biosynthesis and desaturation resulting in an increased fraction of C16 and C18 saturated and during cell differentiation. monounsaturated FAs.

Enhanced Fatty Acid Synthesis Increases Glycerophospholipid Content Phosphatidylcholine and Phosphatidylethanolamine Synthesis Are Fatty During Monocyte Differentiation. Next, we explored lipid composi- Acid Synthesis-Dependent in Macrophages. To show a direct con- tion of cells during differentiation. With 28% of the analyzed lipids, nection of FA and PL synthesis during differentiation of monocytes, cholesterol is the dominating lipid of monocytes followed by we ﬁrst inhibited enzymes required for FA synthesis by block- E phosphatidylcholine (PC) with 24% (Fig. 1 ). Consistent with ing SREBP processing with 25-hydroxycholesterol (25-HC) (11). impaired SREBP-2 target gene expression in macrophages, the free 25-HC dose dependently reduced transcription, protein levels, and cholesterol (FC) fraction declined during cell differentiation to enzyme activities of the SREBP-1c target genes FAS and SCD (Fig. 10%. Moreover, analysis of stable isotope-labeled acetate incor- 2A and Fig. S3). We found that decreased FA synthesis (Fig. 2B)led poration in FC during cell differentiation showed a decreased FC C F to decreased PC (about 65% of control) (Fig. 2 ) and PE synthesis synthesis (Fig. 1 ). The PC content dramatically increased to 40% (about 75% of control) (Fig. 2D). becoming the major lipid class of macrophages. Phosphatidyletha- As a positive control we enhanced FA synthesis by SREBP-1c nolamine (PE + PE-plasmalogens) content was elevated, whereas activation through the LXR-stimulation agent T0901317 (12). phosphatidylserine (PS) levels were lowered. Overall the content of LXR activation increased FA synthesis (Fig. 2B), D -PC (Fig. 2C) PC and ethanolamine-containing phospholipids rose from 40 to 9 and D -PE (Fig. 2D) generation (140% and 120%). However, it almost 70%. Analysis of the species pattern of the individual glyc- 4 was not possible to increase FA and D -PC synthesis of 25-HC- erophospholipid (PL) classes revealed a remarkable shift from the 9 longer and more unsaturated to shorter and less unsaturated PL treated macrophages by LXR stimulation, implying that LXR is species during macrophage differentiation (Fig. S2). not able to induce FAS levels independently from SREBP. α To investigate glycerophospholipid biosynthesis, we supplied cells Moreover, elevated LXR levels observed under 25-HC treatment (Fig. 2A and Fig. S3) could not restore FA synthesis, arguing with stable isotope-labeled choline (D9), ethanolamine (D4), and 13 for an indirect action of LXR on FAS levels via SREBP induction. serine ( C3)for4hand24h(Fig. S1). Monocytes and macrophages at day 1 showed marginal de novo PC and PE synthesis (Fig. 1 G and As 25-HC inhibits both SREBP-1c and SREBP-2 processing and H). By contrast, at day 4 of differentiation D -PC synthesis was thus FA and cholesterol synthesis (11), we next suppressed FA syn- 9 E found increased almost 10-fold and D4-PE synthesis 3-fold. The thesis with cerulenin and C75 (Fig. 2 and Fig. S4), well-charac- species pattern of newly synthesized glycerophospholipids changed terized and speciﬁc inhibitors of FAS (13, 14). Similar to 25-HC, F in a similar way to the total species pattern (Fig. S2). In summary, the cerulenin and C75, reduced D9-PC and D4-PE synthesis (Fig. 2 and increase of PC and PE content during macrophage differentiation G and Fig. S4). Most importantly, the cerulenin- and C75-dependent corresponds very well to the increased glycerophospholipid synthesis decrease of PL synthesis was rescued by addition of a FA-CoA mix rates. The shift to shorter and less unsaturated glycerophospholipid composed of the FAs generated during cell differentiation.

7818 | www.pnas.org/cgi/doi/10.1073/pnas.0912059107 Ecker et al. Downloaded by guest on September 27, 2021 long filopodia (Fig. 3B). Their cytoplasm was packed with organelles including mitochondria, primary lysosomes, long ER sections, and particularly large Golgi complexes (Fig. 3C). When 25-HC was present the cells were smaller with less and shorter filopodia than untreated macrophages, even though they were not as small as monocytes (Fig. 3D). The cells had only a few mitochondria and short ER sections; the Golgi complex, when existent at all, was small. Overall upon addition of 25-HC, the cells appeared rather like monocytes than macrophages. Similarly, suppression of FAS with cerulenin led to smaller cells with fewer and less developed filopodia and organelles (Fig. 3E). After addition of FA-CoAs, 25-HC-treated cells increased their Golgi complex and number of lysosomes; addition of FA-CoAs to cerulenin-treated macrophages led to an increase of filopodia, lysosomes, and mitochondria (Fig. S5). Macrophages, where SREBP-1 and FAS was knocked down with RNAi, had only half of the size of cells treated with a control siRNA (Fig. S6). Taken together, these data show that modulation of FA synthesis during cell differentiation massively affects macrophage ultrastructure. Suppression of FA synthesis prevents development of filopodia and cellular organelles including mitochondria, primary lysosomes, ER, and the Golgi complex.

Induction of Fatty Acid Synthesis Is Crucial for Monocyte Differentiation and Phagocytic Activity of Macrophages. Finally, we asked whether inhibition of lipid synthesis affects monocyte differentiation to macrophages and their key function as phagocytes (2). Chitotriosidase (CHIT1) and human cartilage 39-kDa glycoprotein (CHI3L1) are late macrophage differentiation markers (18, 19). A SREBP-dependent regulation of CHIT1 and CHI3L1 has not yet been described. CHIT1 expression is controlled by the transcription factors C/EBPβ and PU.1; Fig. 1. Increased FA, PC, and PE synthesis during monocyte differentiation. κ fi CHI3L1 is regulated by NF B (20, 21). (A)FAprole during monocyte differentiation, analyzed by GC-MS. (B) As expected, mRNA expression analysis revealed a dramatic Induced FA synthesis in macrophages (d1, d4, and d6). (C) Increased FA desaturation in macrophages (d4 and d6). (D) Increased FA elongation in macro- increase of CHI3L1 and CHIT1 expression during macrophage phages (d1, d4, and d6). (E) Free cholesterol (FC) is the predominant lipid in development (Table S2). Addition of 25-HC massively lowered monocytes, whereas phosphatidylcholine (PC) is the major lipid in macro- CHI3L1 and CHIT1 expression (Fig. 4 A and B). Similarly, inhibition phages (d4 and d6). PC, sphingomyelin (SM), dihydrosphingomyelin (Dih-SM), of FA synthesis with cerulenin and C75 reduced expression levels of phosphatidylethanolamine (PE), PE–based plasmalogenes (PE-pl), phosphati- these late differentiation markers (Fig. 4 A and B and Fig. S4). dylserine (PS), lysophosphatidylcholine (LPC), FC, and cholesteryl ester (CE) Importantly, expression of CHI3L1 and CHIT1 could be restored were quantified by ESI-MS/MS. (F) Decreased cholesterol synthesis in macro- partially in cerulenin- and C75-treated cells by mix of C16 and C18 phages (d1, d4, and d6). (G and H) Increase of D9-PC and D4-PE synthesis during 13 FA-CoAs. To provide more evidence that macrophage differentiation monocyte differentiation. (I and J) After 4 h stable isotope labeling C2-PE and 13 depends on fatty acid synthesis, the macrophage differentiation C3-PS synthesis do not show considerable differences between monocytes and macrophages. (*, P < 0.05; **, P < 0.01; ***, P < 0.001) markers CD11b, CD36, and MRC1 (mannose receptor) were analyzed in cells treated with siRNAs against SREBP-1 or FAS (22–24). Knockdown of FA synthesis with RNAi was accompanied by a low- To further strengthen our findings we next used a gene-specific ered cell-surface expression of CD11b, CD36, and MRC1 (Fig. 4C). targeting approach to knock down fatty acid synthesis. Primary Phagocytic activity of macrophages was investigated by uptake monocytes were transfected with a nontargeting control siRNA or of fluorescent phagobeads. The addition of 25-HC, cerulenin, and siRNAs against SREBP-1 or FAS. qRT-PCR analysis showed a C75 decreased macrophage phagocytosis, respectively (Fig. 4D drop of SREBP-1c and FAS mRNA expression by about 55% and and Fig. S4). Phagocytosis could be rescued partly (25-HC) or 70%, respectively (Fig. 2 H and I and Fig. S6). Both SREBP-1 and completely (cerulenin) when cells were supplemented with a FA- CELL BIOLOGY FAS knockdown reduced fatty acid synthesis to 40% (Fig. 2J) and CoA mix. siRNAs against SREBP-1 and FAS decreased macro- were accompanied by suppressed D9-PC and D4-PE synthesis phage phagocytosis by at least 90% compared to cells transfected (Fig. 2 K and L). with a control siRNA (Fig. 4E). Finally, we explored FA-synthesis- In summary, these data show that during phagocytic differ- dependent uptake of enzymatically modified LDL (E-LDL), a entiation of monocytes the up-regulation of PC and PE synthesis lipoprotein ingested by macrophages through phagocytosis and is clearly linked to FA synthesis. found in advanced atherosclerotic lesions (25). Again, inhibition of FA synthesis resulted in reduced cellular cholesterol levels, indi- Modulation of Fatty Acid Synthesis Alters Macrophage Ultrastructure cating an impaired uptake of E-LDL (Fig. S7). and Organelle Development. Next we asked whether inhibition of In summary, these results show that induction FA synthesis is FA synthesis influences macrophage ultrastructure. It is known absolutely necessary for macrophage differentiation and function. that monocyte differentiation increases the volume of the cells Inhibition of FA synthesis prevents phagocytic capacity of the cells. and the number of intracellular organelles (15–17). Accordingly, electron microscopy showed monocytes as small round cells with Discussion short filapodia and a big nucleus (Fig. 3A). The cytoplasm con- A basic feature of monocytes and a crucial step during athero- tained few primary lysosomes, mitochondria, and short endo- sclerosis development is their ability to differentiate into phag- plasmatic reticulum (ER) sections. In contrast, macrophages (day ocytes. Although several groups investigated alteration of gene 4) had about three times the size of monocytes with numerous and expression associated with monocyte differentiation through

Ecker et al. PNAS | April 27, 2010 | vol. 107 | no. 17 | 7819 Downloaded by guest on September 27, 2021 Fig. 2. Up-regulation of FA synthesis is coupled to PC and PE synthesis (A) Treatment with 25-HC reduces mRNA expression of SREBP-1c target genes FAS and SCD, and increases LXRα transcription in macrophages (d4). (B)FA synthesis is inhibited by 25-HC and induced LXR stimulation (2.5 μM T0901317, 24 h) in macrophages (d4). (C and D)FA synthesis (25-HC)-dependent modulation of PC and PE synthesis. (E) FA synthesis is inhibited by cerulenin in macrophages (d4). (F and G) Inhibition of FA synthesis decreased PC and PE synthesis and is rescued by addition of FA-CoA mix (5 μM FA 16:0-CoA, 2 μM FA 16:1-CoA, 4 μM FA 18:0-CoA, 4 μM FA 18:1-CoA, and 2 μM FA 18:2-CoA for 24 h). (H and I) Decreased SREBP-1c and FAS expression in primary cells treated for 48 h with siRNAs. (J) FA synthesis is inhibited by siRNAs against SREBP-1 and FAS. (K and L) SREBP-1 and FAS- dependent modulation of PC and PE synthesis. (*, P < 0.05; **, P < 0.01; ***, P < 0.001)

global transcriptional profiling approaches (26), the role of lipid 29). Upon activation, its soluble inactive form is translocated to synthesis in this process is poorly understood. Transcriptional the membrane, which is facilitated by electrostatic interactions data of more than two stages of cell differentiation are rare and with negatively charged membrane lipids, for instance FAs. This lipid metabolic and functional data are completely unavailable. conclusion is in good agreement with a study in CHO cells, which We found an inverse transcriptional regulation of SREBP-1c and showed that increased PC synthesis is mainly due to induced fatty SREBP-2 target genes leading to strongly induced fatty acid synthesis acid synthesis rather than a transcriptional activation of CCT (30). and a decreased cholesterol fraction during phagocytic differ- The down-regulation SREBP-2 target genes during phagocytic entiation of monocytes. Besides decreased cholesterol synthesis, the differentiation of monocytes was accompanied by a decreased FC lower cholesterol fraction during cell differentiation might also be fraction. Together with an increase in PL content due to enhanced due to LXR-mediated induction of the cholesterol transporters de novo synthesis the FC/PL ratio dramatically dropped from 0.5 ABCA1 and ABCG1 (Table S2) and increased cholesterol efflux. in monocytes to 0.1 in macrophages. Cholesterol is one of the most Extensive characterization of monocyte and macrophage lipid important regulators of lipid organization decreasing membrane metabolism showed that up-regulation of FA synthesis during cell fluidity (31). Thus, a lower FC/PL ratio in macrophages results in differentiation is the driving force for enhanced PC and PE synthesis. more fluid and flexible membranes that support phagocytosis. Importantly, induction of FA and phospholipid synthesis is inde- Therefore PL synthesis is absolutely necessary for macrophage pendent of serum in the macrophage cultivation media, because function, which is also confirmed by the finding that cytokine addition of serum during differentiation does not decrease lipid secretion of macrophages requires PC synthesis (32). synthesis (Fig. S8). More than half of the PE-containing lipids in macrophages are Because we did not observe a transcriptional increase of genes PE-pl, which are known to protect macrophages from oxidative directly involved in PC and PE biosynthesis (Table S2), either a stress by their ability to bind and scavenge free radicals (33). posttranscriptional regulation of these enzymes or a metabolic Concerning cellular damage, a low FC/PL ratio in macrophages activation (through an enlarged fatty acid pool) of PL synthesis may also be an adaptive precaution of cells to protect from cho- may be concluded. Choline-phosphate cytidyltransferase (CCT) lesterol-mediated cytotoxicity during uptake of modified lip- (Fig. S1, PCYT1), the rate-limiting enzyme of de novo PC syn- oproteins. High free cholesterol levels are known to form cytotoxic thesis, is mainly regulated on the posttranscriptional level (27– cholesterol crystals (34). Hence, mature macrophages may build a

7820 | www.pnas.org/cgi/doi/10.1073/pnas.0912059107 Ecker et al. Downloaded by guest on September 27, 2021 Fig. 4. Induction of FA synthesis is crucial for phagocytic differentiation and Fig. 3. Modulation of FA synthesis alters macrophage ultrastructure and function. (A and B) Suppression of CHI3L1 and CHIT1 expression by inhibition organelle development. Ultrastructural analysis was performed with trans- of FA synthesis and rescue by addition of FA-CoA mix, analyzed with qRT- mission electron microscopy. (Scale bar, 0.5 μm.) Endoplasmic reticulum (ER), PCR. (C) Decreased surface expression of CD11b, CD36, and MRC1 (CD206) of lysosome (Ly), mitochondrium (M), nucleus (N). A, B, D,andE are scaled equally. primary macrophages treated for 48 h with siRNAs, analyzed by flow Relative sizes and organelle content were estimated from multiple EM pictures cytometry. (D) Inhibition of FA synthesis reduced phagocytosis of macro- for each condition. (A) Monocytes are small round cells with short filopodia and phages (d4), which is rescued by addition of FA-CoA mix, analyzed by flow a big centered nucleus. (B) Macrophages (d4) have about three times the size of cytometry. (E) Knockdown of SREBP-1 and FAS suppresses phagocytosis of monocytes, containing numerous and long filopodia and their cytoplasm is primary macrophages. (*/#, P < 0.05; **/##, P < 0.01; ***/###, P < 0.001. *, P packed with organelles (Golgi, ER, Ly, and M massively increase). (C)Magnifi- values were calculated relative to untreated samples; #, P values were cal- cation from B highlighting the large Golgi complexes of macrophages. (D) culated relative to cerulenin- or 25-HC-treated samples.) Inhibition of lipid synthesis [2.5 μM 25-HC, Mac(d4)] decreases cell size (by ∼40%), filopodia, and cellular organelles (by ∼70% for M, by ∼80% for Ly; Golgi is missing, only short ER sections). (E) Inhibition of FAS [2 μg/mL cerulenin, Mac scriptional induction of the scavenger receptor CD36 (5) and is ∼ fi ∼ (d4)] decreases cell size (by 40%), lopodia, and cellular organelles (by 50% required macrophage activation (4, 41). This might also explain our for M, by ∼70% for Ly, by ∼50% for Golgi, short ER sections). finding that the uptake of modified lipoproteins is inhibited when fatty synthesis is suppressed during the differentiation process. “buffer” for further cholesterol uptake, with an enhanced PL In summary, we could identify an inverse regulation of FA/ content through high PL synthesis. Blocking of PC synthesis during phospholipid and FC synthesis during phagocytic differentiation FC loading results in enhanced death of macrophages (35). of primary human monocytes leading to a decreased FC/PL Analyzing cell morphology during cell differentiation showed that ratio. This adaptation of lipid metabolism is required for cellular macrophages strongly increase their size and number of intracellular organelle development and macrophage function. Consequently, organelles including the Golgi complex, ER, and primary lysosomes. our results suggest a previously unknown role for FA synthesis in We found that modulation of fatty acid synthesis massively impacts the phagocytic differentiation of monocytes, which may be fi CELL BIOLOGY cell size, lopodia, and cellular organelle development. In mam- important for tissue development, host defense, and in human malian cells, the FC/PL ratio of the plasma membrane is 1.0, whereas metabolic diseases such as atherosclerosis. membranes of the Golgi complex and the ER have a FC/PL ratio of 0.2 and 0.15, and the FC/PL ratio of lysosomal membranes is 0.38 (36, Materials and Methods 13 13 37). Thus these organelles have very low amounts of cholesterol in Reagents and Materials. The 2- C-acetate, C3-serine, D3-palmitate, D4-

their membranes; their main membrane lipids are the phospholipids ethanolamine, and D9-choline were obtained from Cambridge Isotope PC and PE (36), which fits very well with the up-regulation of Laboratories. 25-HC, cerulenin, C75, T0901317, and fatty acid-CoAs were phospholipid synthesis during cell differentiation and the change or purchased from Sigma-Aldrich. their lipid composition. Importantly, the ER is a source of membrane for phagosome formation and therefore expansion of ER is a pre- Monocyte Isolation and Cell Culture. Primary human monocytes were obtained fl requisite for phagocytosis in macrophages (38, 39). In our experi- from healthy donors by leukapheresis and counter ow elutriation as described previously (42). Cells were cultured on plastic Petri dishes in macrophage SFM ments inhibition of FA synthesis resulted in less ER, which might also medium (Gibco-BRL) and allowed to differentiate for 1, 4, or 6 days in the contribute to a reduced phagocytosis in these cells. presence of 50 ng/mL recombinant human MCSF from R&D Systems. Beside biophysical and metabolic effects, an increased FA pool may also influence nuclear hormone receptor activity in macro- RNA Isolation and Quantitative RT-PCR Analysis. Total RNA was extracted with phages. PPARs are well-known sensors of fatty acids (40). Induction the RNeasy Midi Kit (Qiagen); cDNA was generated using the Reverse Tran- of PPARγ promotes the uptake of oxidized LDL through tran- scripton System from Promega. Real-time quantitative RT-PCR analysis was

Ecker et al. PNAS | April 27, 2010 | vol. 107 | no. 17 | 7821 Downloaded by guest on September 27, 2021 performed with an ABI7900HT machine (Applied Biosystems). Transcripts have internal standards. Lipids were quantified by electrospray ionization tandem been specified with predesigned and optimized Assays on Demand (Applied mass spectrometry (ESI-MS/MS) in positive ion mode as described previously Biosystems); 18S rRNA was used as reference gene. Relative quantification was (44) (for details, see Fig. S1). carried out with the SDS 2.3 software (Applied Biosystems). Ultrastructural Analysis. Ultrastructural analysis was performed with trans- Protein Isolation and Analyses. Total cell lysates were prepared in RIPA buffer missionelectronmicroscopywithstandardprotocolsasdescribedpreviously(45). (Roche). Samples with equal amounts of protein were separated on SDS gels, β transferred on PVDF membranes, and incubated with anti- -actin (Sigma- Phagocytosis Assays and Other FACS Analysis. Phagocytic capacity was Aldrich), anti-FAS (Assay Design), anti-SREBP-1 antibodies (Santa Cruz) with a determined with flow cytometry. Primary macrophages were incubated with dilution of 1:1,000 for 2 h. Peroxydase-conjugated goat anti-mouse IgG 1.1 × 1011 phagobead particles/mL (fluoresbrite yellow green microspheres, (Dianova) were used at 1:20,000 for 1 h and visualized with chemo- 0.75 μm, Polysciences Europe GmbH); 2 h later, fluorescence was determined luminescence on film (ECL Plus, Amersham Pharmacia Biotech). with a FACS Calibur flow cytometer (BD Biosciences). Expression of CD11b, CD36, and MRC1 (CD206) was determined by flow cytometry (1 × 105 cells RNAi Experiments. Freshly isolated human monocytes were obtained from per analysis) with antibodies from BD Biosciences. healthy volunteers; subsequently 1 million cells were electroporated with 50 nM siRNAusingtheHumanMonocyteNucleofectorKit(Amaxa)andseededin6-well Statistics. The level of significance for the difference between data sets was plates with macrophage medium containing MCSF for 48 h. All siRNAs were assessed using Student’s independent t test (*/#, P < 0.05; **/##, P < 0.01; “human validated siRNAs” from Ambion; the sequence of the nontargeting ***/###, P < 0.001). *, P values were calculated relative to control samples; #, control siRNA was 5′-AGUACUGCUUACGAUACGGTT-3′, the sequence of the – SREBP-1 siRNA was 5′-GGCAAAGCUGAAUAAAUCUTT-3′, and the sequence of P values were calculated relative to cerulenin- or 25-HC treated samples. ± the FAS siRNA was 5′-GGUAUGCGACGGGAAAGUATT-3′. Results are expressed as mean SEM.

Fatty Acid Analysis. Total FA analysis was carried out by GC-MS (for details, see ACKNOWLEDGMENTS. This work was supported by the seventh framework program of the European Union-funded “LipidomicNet” (Proposal 202272) Fig. S1). and SFB-TR 13/A3. We thank Jolante Aiwanger, Manfred Haas, Doreen Mül- ler, Simone Peschel, Birgit Wilhelm, Christina Köppler, and especially Barbara Lipid Analysis. Lipids were extracted according to the procedure described by Tille for excellent technical assistance, and Gerrit van Meer for helpful sci- Bligh and Dyer (43) in the presence of nonnaturally occurring lipid species as entiﬁc discussions.

1. Pollard JW (2009) Trophic macrophages in development and disease. Nat Rev 24. Gazi U, Martinez-Pomares L (2009) Influence of the mannose receptor in host immune Immunol 9:259–270. responses. Immunobiology 214:554–561. 2. Hume DA (2006) The mononuclear phagocyte system. Curr Opin Immunol 18:49–53. 25. Bhakdi S, et al. (1995) On the pathogenesis ofatherosclerosis: Enzymatic transformationof 3. Hamilton JA (2008) Colony-stimulating factors in inflammation and autoimmunity. human low density lipoprotein to an atherogenic moiety. JExpMed182:1959–1971. Nat Rev Immunol 8:533–544. 26. Hume DA (2008) Differentiation and heterogeneity in the mononuclear phagocyte 4. Bouhlel MA, et al. (2007) PPARgamma activation primes human monocytes into system. Mucosal Immunol 1:432–441. alternative M2 macrophages with anti-inflammatory properties. Cell Metab 6:137–143. 27. Attard GS, Templer RH, Smith WS, Hunt AN, Jackowski S (2000) Modulation of CTP: 5. Tontonoz P, Nagy L, Alvarez JG, Thomazy VA, Evans RM (1998) PPARgamma promotes phosphocholine cytidylyltransferase by membrane curvature elastic stress. Proc Natl monocyte/macrophage differentiation and uptake of oxidized LDL. Cell 93:241–252. Acad Sci USA 97:9032–9036. 6. Horton JD, Goldstein JL, Brown MS (2002) SREBPs: Activators of the complete program 28. Cornell RB, Northwood IC (2000) Regulation of CTP:phosphocholine cytidylyltransferase of cholesterol and fatty acid synthesis in the liver. J Clin Invest 109:1125–1131. by amphitropism and relocalization. Trends Biochem Sci 25:441–447. 7. Smith S, Witkowski A, Joshi AK (2003) Structural and functional organization of the 29. Vance JE, Vance DE (2004) Phospholipid biosynthesis in mammalian cells. Biochem – animal fatty acid synthase. Prog Lipid Res 42:289–317. Cell Biol 82:113 128. 8. Matsuzaka T, et al. (2007) Crucial role of a long-chain fatty acid elongase, Elovl6, in 30. Ridgway ND, Lagace TA (2003) Regulation of the CDP-choline pathway by sterol obesity-induced insulin resistance. Nat Med 13:1193–1202. regulatory element binding proteins involves transcriptional and post-transcriptional – 9. Miyazaki M, Ntambi JM (2003) Role of stearoyl-coenzyme A desaturase in lipid mechanisms. Biochem J 372:811 819. fi metabolism. Prostaglandins Leukot Essent Fatty Acids 68:113–121. 31. Max eld FR, Tabas I (2005) Role of cholesterol and lipid organization in disease. – 10. Matsuzaka T, et al. (2002) Dual regulation of mouse Delta(5)- and Delta(6)-desaturase Nature 438:612 621. gene expression by SREBP-1 and PPARalpha. J Lipid Res 43:107–114. 32. Tian Y, et al. (2008) Cytokine secretion requires phosphatidylcholine synthesis. J Cell – 11. Adams CM, et al. (2004) Cholesterol and 25-hydroxycholesterol inhibit activation of Biol 181:945 957. SREBPs by different mechanisms, both involving SCAP and Insigs. J Biol Chem 279: 33. Reiss D, Beyer K, Engelmann B (1997) Delayed oxidative degradation of polyunsaturated diacyl phospholipids in the presence of plasmalogen phospholipids in vitro. Biochem J 323: 52772–52780. 807–814. 12. Yoshikawa T, et al. (2001) Identification of liver X receptor-retinoid X receptor as an 34. Tabas I (2002) Consequences of cellular cholesterol accumulation: Basic concepts and activator of the sterol regulatory element-binding protein 1c gene promoter. Mol Cell physiological implications. J Clin Invest 110:905–911. Biol 21:2991–3000. 35. Zhang D, et al. (2000) Macrophages deficient in CTP:Phosphocholine cytidylyltransferase- 13. Moche M, Schneider G, Edwards P, Dehesh K, Lindqvist Y (1999) Structure of the alpha are viable under normal culture conditions but are highly susceptible to free complex between the antibiotic cerulenin and its target, beta-ketoacyl-acyl carrier cholesterol-induced death. Molecular genetic evidence that the induction of protein synthase. J Biol Chem 274:6031–6034. phosphatidylcholine biosynthesis in free cholesterol-loaded macrophages is an adaptive 14. Kuhajda FP, et al. (2000) Synthesis and antitumor activity of an inhibitor of fatty acid response. JBiolChem275:35368–35376. synthase. Proc Natl Acad Sci USA 97:3450–3454. 36. van Meer G, Voelker DR, Feigenson GW (2008) Membrane lipids: Where they are and 15. Sutton JS, Weiss L (1966) Transformation of monocytes in tissue culture into how they behave. Nat Rev Mol Cell Biol 9:112–124. macrophages, epithelioid cells, and multinucleated giant cells. An electron 37. Schoer JK, et al. (2000) Lysosomal membrane cholesterol dynamics. Biochemistry 39: – microscope study. J Cell Biol 28:303 332. 7662–7677. fi 16. Krombach F, et al. (1996) Characterization and quanti cation of alveolar monocyte- 38. Gagnon E, et al. (2002) Endoplasmic reticulum-mediated phagocytosis is a mechanism fl – like cells in human chronic in ammatory lung disease. Eur Respir J 9:984 991. of entry into macrophages. Cell 110:119–131. 17. Sokol RJ, Hudson G, James NT, Frost IJ, Wales J (1987) Human macrophage 39. Becker T, Volchuk A, Rothman JE (2005) Differential use of endoplasmic reticulum – development: A morphometric study. J Anat 151:27 35. membrane for phagocytosis in J774 macrophages. Proc Natl Acad Sci USA 102:4022–4026. 18. Boot RG, Renkema GH, Strijland A, van Zonneveld AJ, Aerts JM (1995) Cloning of a 40. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ (2001) Nuclear receptors and lipid cDNA encoding chitotriosidase, a human chitinase produced by macrophages. J Biol physiology: Opening the X-files. Science 294:1866–1870. – Chem 270:26252 26256. 41. Odegaard JI, et al. (2007) Macrophage-specific PPARgamma controls alternative fi 19. Krause SW, et al. (1996) Differential screening identi es genetic markers of monocyte activation and improves insulin resistance. Nature 447:1116–1120. to macrophage maturation. J Leukoc Biol 60:540–545. 42. Ecker J, Langmann T, Moehle C, Schmitz G (2007) Isomer specific effects of conjugated 20. Pham TH, et al. (2007) CCAAT enhancer-binding protein beta regulates constitutive linoleic acid on macrophage ABCG1 transcription by a SREBP-1c dependent gene expression during late stages of monocyte to macrophage differentiation. J Biol mechanism. Biochem Biophys Res Commun 352:805–811. Chem 282:21924–21933. 43. Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification. 21. Recklies AD, Ling H, White C, Bernier SM (2005) Inflammatory cytokines induce Can J Biochem Physiol 37:911–917. production of CHI3L1 by articular chondrocytes. J Biol Chem 280:41213–41221. 44. Binder M, Liebisch G, Langmann T, Schmitz G (2006) Metabolic profiling of 22. Mazzone A, Ricevuti G (1995) Leukocyte CD11/CD18 integrins: Biological and clinical glycerophospholipid synthesis in fibroblasts loaded with free cholesterol and relevance. Haematologica 80:161–175. modified low density lipoproteins. J Biol Chem 281:21869–21877. 23. Collot-Teixeira S, Martin J, McDermott-Roe C, Poston R, McGregor JL (2007) CD36 and 45. Robenek H, et al. (2006) Adipophilin-enriched domains in the ER membrane are sites macrophages in atherosclerosis. Cardiovasc Res 75:468–477. of lipid droplet biogenesis. J Cell Sci 119:4215–4224.

7822 | www.pnas.org/cgi/doi/10.1073/pnas.0912059107 Ecker et al. Downloaded by guest on September 27, 2021